Select a calendar:
Filter May Events by Event Type:
SUNMONTUEWEDTHUFRISAT
Events for the 5th week of May
-
PhD Dissertation Defense - Myrl Marmarelis
Tue, May 28, 2024 @ 02:00 PM - 04:00 PM
Thomas Lord Department of Computer Science
University Calendar
Title: Robust Causal Inference with Machine Learning on Observational Data
Date and Time: Tuesday, May 28th - 2:00pm - 4:00pm
Committee: Aram Galstyan (Chair), Greg Ver Steeg, Fred Morstatter, Shanghua Teng, and Roger Ghanem (external)
Abstract:
The rise of artificial intelligence and deep learning has led to unprecedented capabilities in prediction. As these black-box algorithms are deployed in different parts of society, it is becoming increasingly clear that predictions alone do not always translate to enabling effective decisions, policies, or reliable forecasts in a changing world. What is often needed is a stronger understanding of a system than a predictive model of observations can offer. This deficit arises when attempting to predict the system’s behavior in novel situations. Causal inference refers to a set of theoretical frameworks and practical methods for identifying cause-and-effect structures from data. Knowledge of this structure can help anticipate what would happen in a novel situation, like subjecting the system to intervention. Much work in causal inference is concerned with finding the minimal assumptions required to answer specific causal questions, like estimating the effect of a certain treatment. The more reasonable and relaxed the assumptions of a causal-inference method, the more applicable it is to diverse datasets and machine learning. There are many methodological aspects to performing causal inference on observational data—that is, without the ability to perform experiments. Of fundamental significance is having workable representations of the system that can be learned from data. Closely related to the quality of the representations is the ability to make downstream causal estimates robust to confounding. Confounders in a system are common structures that might confuse apparent relations between cause and effect, or treatment and outcome.
In this dissertation, I propose methods for addressing these problems in challenging machine-learning contexts. I introduce an improved representation of single-cell RNA sequencing data for inference tasks in medicine and biology. Looking for high-dimensional interactions in biological processes leads to better resolution of phenotypes. More broadly, I make numerous contributions towards increased robustness of machine learning to hidden or observed confounding. I address sensitivity of dose-response curves to hidden confounding, prediction of interventional outcomes under hidden confounding; robust effect estimation for continuous-valued and multivariate interventions, and estimation for interventions that might only encourage treatment as a function of susceptibility.
Location: Information Science Institute (ISI) - 553
Audiences: Everyone Is Invited
Contact: Myrl Marmarelis
-
MHI Seminar - Karen Livescu - Tuesday, May 28th at 3pm in EEB 248 & Zoom
Tue, May 28, 2024 @ 03:00 PM - 05:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Karen Livescu, Professor TTI-Chicago
Talk Title: What Do Pre-Trained Speech Representation Models Know?
Abstract: Pre-trained speech representation models have become ubiquitous in speech processing over the past few years. They have both improved the state of the art and made it feasible to learn task-specific models with very little labeled data. However, it is not well understood what linguistic information is encoded in pre-trained models, where in the models it is encoded, and how best to apply this information to downstream tasks. In this talk I will describe recent work that begins to build an understanding of pre-trained speech models, through both layer-wise analysis and benchmarking on tasks. We consider a number of popular pre-trained models and investigate the extent to which they encode spectral, phonetic, and word-level information. The results of these analyses also suggest some ways to improve or simplify the application of pre-trained models for downstream tasks. Finally, I will describe our efforts to benchmark model performance on a variety of spoken language understanding tasks, in order to broaden our understanding of the semantic capabilities of speech models.
Biography: Karen Livescu is a Professor at TTI-Chicago. This year she is on sabbatical, splitting her time between the Stanford NLP group and the CMU Language Technologies Institute. She completed her PhD at MIT in 2005. She is an ISCA Fellow and a recent IEEE Distinguished Lecturer. She has served as a program chair/co-chair for ICLR, Interspeech, and ASRU, and is an Associate Editor for TACL and IEEE T-PAMI. Her group's work spans a variety of topics in spoken, written, and signed language processing, with a particular interest in representation learning, cross-modality learning, and low-resource settings.
Host: Shrikanth Narayanan
More Info: https://usc.zoom.us/j/98343896109?pwd=VWxRVTJVc3NLMjZGcEVVNGw1a1J0dz09
More Information: 2024 Karen Livescu Seminar.pdf
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited
Contact: Marilyn Poplawski
Event Link: https://usc.zoom.us/j/98343896109?pwd=VWxRVTJVc3NLMjZGcEVVNGw1a1J0dz09
-
PhD Thesis Proposal - Siyi Guo
Wed, May 29, 2024 @ 12:00 PM - 01:30 PM
Thomas Lord Department of Computer Science
University Calendar
Title: Understanding Population Heterogeneities through Dynamic Behaviors
Committee: Kristina Lerman (Chair), Fred Morstatter, Urbashi Mitra, Shanghua Teng
Location: SAL 322
Date and Time: Weds., May 29th: 12:00p - 1:30p
Abstract:
The rich and dynamic information environment of social media provides researchers, policy makers, and entrepreneurs with opportunities to learn about social phenomena in a timely manner. However, using these data to understand social behavior is difficult due to the long-tailed distributions of both contents and user attributes and the heterogeneity of topics and events discussed in the highly dynamic online environment. Existing methods typically rely on specific features like text content, activity patterns, or platform metadata, failing to holistically model user behavior across different modalities. To address these challenges, we aim to discover and model population heterogeneities by studying user behavioral dynamics on social media. First, we present a method for systematically detecting and measuring emotional reactions to offline events, and use it to uncover the different emotional reactions in US liberal and conservative populations to the overturn of Roe v. Wade. In the second part, we further model the heterogeneous user behaviors by a novel social media user representation learning framework, and demonstrate its versatility through two applications: 1) Measuring increased polarization in online discussions after major events by quantifying how users with different beliefs moved farther apart in the embedding space, and (2) Identifying inauthentic accounts involved in coordinated influence operations by detecting users posting similar content simultaneously. Our ability to discover and model user heterogeneity enables new solutions to important problems around disinformation, societal tensions, and online behavior understanding.
Zoom: https://usc.zoom.us/my/siyiguoLocation: Henry Salvatori Computer Science Center (SAL) - 322
Audiences: Everyone Is Invited
Contact: Siyi Guo
Event Link: https://usc.zoom.us/my/siyiguo
-
PhD Dissertation Defense - Xin Qin
Fri, May 31, 2024 @ 09:30 AM - 11:30 PM
Thomas Lord Department of Computer Science
University Calendar
Presentation title: Data-driven and Logic-based Analysis of Learning-enabled Cyber-Physical Systems
Names of the guidance committee members: Jyotirmoy Deshmukh, Chao Wang, Souti Chattopadhyay, and Yan Liu
Abstract:
Rigorous analysis of cyber-physical systems (CPS) is becoming increasingly important, particularly for safety-critical applications incorporating learning-enabled components. Given a system requirement such as "if the system deviates from the center of the road, it should return to the center in time," we aim to evaluate how well the system satisfies this requirement in uncertain environments. The defense will center around three main pillars: (1) performing verification for initial states and during the runtime of the system, (2) demonstrating how to reuse verification results for unseen systems, and (3) designing new specification languages to alleviate sensitivity to noise. Since these three pillars all involve a similar approach of black-box modeling and analysis using properties related to specification languages, we anticipate that future work could integrate the results from various stages of this thesis. This integration would facilitate the sharing and reuse of findings at each stage, thereby enhancing system safety analysis and improving the scalability of the reasoning process.
Location: Henry Salvatori Computer Science Center (SAL) - 213
Audiences: Everyone Is Invited
Contact: Ellecia Williams
-
Human-AI Interaction: From Supporting Surgical Training to Inspecting Social Bias in LLMs
Fri, May 31, 2024 @ 11:00 AM - 12:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Rafal Kocielnik, California Institute of Technology
Talk Title: Human-AI Interaction: From Supporting Surgical Training to Inspecting Social Bias in LLMs
Series: AI Seminar
Abstract: *Meeting hosts only admit on-line guests that they know to the Zoom meeting. Hence, you’re highly encouraged to use your USC account to sign into Zoom. If you’re an outside visitor, please inform us at (aiseminars-poc(at)isi.edu) to make us aware of your attendance so we can admit you. Specify if you will attend remotely or in person at least one business day prior to the event Provide your: full name, job title and professional affiliation and arrive at least 10 minutes before the seminar begins. If you do not have access to the 6th Floor for in-person attendance, please check in at the 10th floor main reception desk to register as a visitor and someone will escort you to the conference room location.
In this talk, I will present my recent contributions to Human-AI interaction, focusing on two distinct projects looking at opportunities and challenges involved in the use of modern AI. In the first part of my talk, I will present my work on leveraging AI in clinician education, specifically within the surgical context. I will detail my work on utilizing multimodal deep-learning techniques to analyze formative feedback from surgeons to trainees in the context of real-world robot-assisted surgeries. This project marks a significant step forward in harnessing contemporary AI for the specialized domain of surgical education, receiving the best paper award at the ML4H conference. For the second part of my talk, I will focus on Human-AI interaction in the context of empowering domain experts (e.g., social scientists and ethicists) to inspect modern generative AI for the presence of harmful stereotypes. I will describe our BiasTestGPT framework which offers two important contributions: 1) a novel approach for generating high-quality synthetic data for social bias testing at scale and 2) a user-friendly and open-sourced interface for engaging the general public and domain experts in the inspection of modern AI. Together, these projects demonstrate opportunities in leveraging Human-AI interaction for supporting specialized domains and helping inspect the challenges in AI itself. This event will be recorded. It will be posted on our USC/ISI YouTube page within 1-2 business days: https://www.youtube.com/user/USCISI.
Biography: RafaÅ Kocielnik is a Postdoctoral Researcher at Caltech's Computing+Mathematical Sciences department, where he also collaborates with Cedars-Sinai Medical Center and Activision Blizzard gaming company. He holds an M.Sc. in Computer Science from the Polish-Japanese Academy of Information Technology, a P.D.Eng. in Industrial Design from Eindhoven University of Technology and completed his Ph.D. in Human-Centered Design & Engineering at the University of Washington, Seattle, in 2021. His focus was on designing engaging conversational interactions for health and behavior change. Awarded a CRA Computing Innovation Fellowship in 2021, his research at Caltech explores the intersection of AI and HCI with applications in surgical training, social bias testing in Generative AI, and toxicity mitigation in gaming. He has received Best Paper awards at CSCW and ML4H, with an Honorable Mention at CUI, underscoring his interdisciplinary focus and commitment to advancing AI and HCI for human-centered applications. Visit links below to subscribe and for details on upcoming seminars: https://www.isi.edu/isi-seminar-series/ https://www.isi.edu/events/
Host: Myrl Marmarelis and Justina Gilleland + Maura Covaci
More Info: https://www.isi.edu/events/4976/human-ai-interaction-from-supporting-surgical-training-to-inspecting-social-bias-in-llms/
Webcast: https://usc.zoom.us/j/99601436181?pwd=d0Y5eTZPbHRjM2t3NHc5cXRMNkE1dz09Location: Information Science Institute (ISI) - Conf Rm#1135-1137
WebCast Link: https://usc.zoom.us/j/99601436181?pwd=d0Y5eTZPbHRjM2t3NHc5cXRMNkE1dz09
Audiences: Everyone Is Invited
Contact: Pete Zamar
